Biostatistics I - PH 395 OL1 - Course Syllabus
Content
The University of Vermont Biostatistics 1 (Applied Research Methods in Public Health) Syllabus Name: PH 303: Biostatistics I (Applied Research Methods in Public Health) Structure: On-line course Credits: 3 Faculty: Thomas Delaney, PhD Research Associate, Department of Pediatrics UVM College of Medicine, VCHIP UHC St. Joseph 7; Room 7211 1 South Prospect St., Burlington, VT 05401 Phone: 802-656-9192 Email: [email protected] Competencies—Attached. Course overview: The course is based on real world applied research needs and how those needs can best be met. The course consists of discussion and critiques of published articles, presentations by the instructor and working with example qualitative and quantitative data sets. The course will be taught online and be divided into 12 modules. The instructor will periodically post PowerPoint slide sets that will be reviewed on-line prior to specific modules in the course. Requirements include: brief (2 page) critiques of articles and situations presented by the instructor, mandatory participation in chat/blog based discussions related to each module from the course, and developing a Research Proposal for an applied research project and that will be shared as part of the final module. Each module has a specific reading(s) associated with it and that will have to be done in advance of the beginning of the module. Additional requirements for completion prior to each module are described below in the course calendar. Course assignments and grading: For each module in the course, students will need to complete the required readings, critiques and other work prior to the scheduled on-line chat/blog sessions for that module. Grades will be based on class participation via discussion board postings (40%), brief (approximately two pages each) critiques of research articles (20%), completion of five problem sets that focus on applying statistical and methodological learning (20%) and the development and presentation of the Research Proposal for an applied research project (20%). The course is designed to be taught as either a one month (four week) course (three modules per week) or a semester long course. Discussion board postings (40% of final grade): For each module, there will be a discussion thread that is started by the instructor, and in which all students must
1
participate. There is no minimum number of times students must participate, but contributions totaling at least 200 words will be expected. Postings will be evaluated for the quality of their content and how well they reflect the readings and/or PowerPoint presentations for each module. Each discussion will be open for one week. Article critiques (20% of final grade): A total of seven articles will be read during the course, and for each article students will write an approximately two page “critique” of the article. Critiques should be written clearly and reflect the students’ learning from the corresponding course module as well as from previous modules. Students will not be graded on spelling or punctuation but rather on how well their writing reflects key ideas from the course and based on their application of critical thinking skills. Problem sets (20% of final grade): A total of five statistical/methodological problem sets will be completed during the course. Each problem set will provide students with background information about an applied public health dataset and the data itself. Each problem set will also describe a specific question that students will answer, using a combination of what they have learned in the corresponding (and previous) modules and the SPSS v18 statistical software package (available for free as a download to UVM students). Students’ work for each problem set will be evaluated based on the quality of a brief (one page or less) written summary of the answers to the questions posed for the problem, as well as copies of the statistical output that supports their answers. Final research proposal (20% of final grade): Starting during Module 4, students will begin developing a sample research proposal that addresses an applied public health research question. For example, a student might write a mock grant proposal to do a study of the effects of fluoridation on a set of dental health measures in a rural community, in which they described the nature of the research question, propose methods for answering the questions and describe how they would collect and analyze data that would allow them to make a strong inference about the applied research question. Proposals will be between seven and ten pages in length, and will be evaluated based on how well the proposal demonstrates mastery of the concepts presented in the course. Readings: The main text for the course is Essentials of Biostatistics in Public Health (2nd edition, 2011) by Lisa Sullivan. Additional readings will consist of journal articles and book chapters, as well as PowerPoint presentations developed by the instructor, and these are listed as part of each module described below. Readings should be completed prior to the start of each module.
Course calendar: Module 1 Description: Overview of course and requirements. What is applied public health research, and how does it differ from other sorts of research? Examples of applied public health research
2
designs. What sorts of questions does applied research in public health answer, and what needs does it meet? Review of studies that represent rigorous applied studies. Readings: Sullivan, Chapter 2. Instructor PowerPoint presentation. Module 2 Description: What are validity and reliability? How do we assess validity and reliability or data collection procedures, instruments, analyses and claims? What are threats to validity and reliability of applied research, and how can these be ameliorated? What are the biases that exist in applied research, and how can these be minimized? Readings: Bowling, A (2002) Research methods in health: investigating health and health services. (Chapter 6, pages 144-153). Levin, S et al. (2003) Physical Activity and Body Mass Index Among US Adolescents. Arch Pediatr Adolesc Med.157:816-820.
Critique of:
Module 3 Description: More detailed exploration of the types of data that are available for applied public health research. Using existing data sets versus original data collection. Issues around data quality and proper data management. Review of how data should be organized for analysis. Readings: Sullivan, Chapter 4. Instructor PowerPoint presentation. Problem set: Identifying threats to data validity and errors in data management in a sample data set. Module 4 Description: Review of qualitative approaches to data collection, including focus (discussion) groups, interviews and others. Identifying appropriate questions and methods for collecting these data. Analyzing qualitative data using less formal versus more formal approaches. Design of effective surveys and interview guides. Description and discussion of the requirement for the Research Proposal course requirement. Readings: Jones J and Hunter D (1995) Qualitative Research: Consensus methods for medical and health services research. BMJ 311 : 376 (Published 5 August 1995) Instructor PowerPoint presentation.
3
Critique of:
Crowe, M et al. (2011) Was it something I did wrong? A qualitative analysis of parental perspectives of their child's bipolar disorder. Journal of Psychiatric and Mental Health Nursing: 18:4, 342–348.
Module 5 Description: Overview of quantitative data sources, including behavioral, biological and environmental data. Outcome versus process measures. Implications of the type and quality of data for the strength of the descriptions/ inferences you are trying to make. Overview of probability distributions. Readings: Sullivan, Chapter 5. Instructor PowerPoint presentation.
Module 6 Description: Descriptive (quantitative) data analysis. What statistics are used for descriptive quantitative research, and what do they tell us? Example descriptive analyses using small data sets and the SPSS software package. Readings: Sullivan, Chapter 6. Instructor PowerPoint presentation. Problem set: Using SPSS v18 to describe a dataset that includes demographic and outcome variables.
Module 7 Description: Inferential (quantitative) data analysis—Part 1. Statistical approaches for small and/or highly variable datasets. Overview of non-parametric versus traditional statistical analysis, and their relative strengths and weaknesses. Analyzing categorical data sets. Reviewing published articles with an eye to the validity of their statistical analysis. Readings: Critique of: Sullivan, Chapter 10. Duncan, P et al. (unpublished). A Statewide Initiative to Support Primary Care Practices in Improving Adolescent Preventive Services.
Problem set: Using SPSS v18 to conduct hypothesis testing using categorical statistical tests such as Fisher’s Exact test and Chi Square.
4
Module 8 Description: Inferential (quantitative) data analysis—Part 2. Overview of statistical approaches for more robust quantitative data sets, including univariate (e.g., t-tests, ANOVA) and multivariate (e.g., logistic regression, MANOVA) analysis. Reviewing published articles with an eye to the validity of their statistical analysis. Readings: Sullivan, Chapters 7 & 9.
Problem set: Using SPSS v18 to conduct univariate hypothesis testing with T-tests, ANOVA and non-parametric (e.g., Kruskall-Wallace) statistical tests.
Module 9 Description: Overview of effective presentation of quantitative data. Why are some data displays more versus less appropriate? Examples of effective data displays, and discussion of how to reduce the potential for confusing consumers of quantitative reports. Readings: Critique of: Instructor PowerPoint presentation. van Heuvelen, M et al. (2005) Differences between participants and nonparticipants in an RCT on physical activity and psychological interventions for older people. Aging Clin Exp Res. 17; 236-245.
Module 10 Description: Program evaluation as applied public health research: goals, methods, analysis and reporting of results. How are program evaluation data different from other data? Review of example program evaluations and discussion of the relative strengths and barriers to effective program evaluation. Readings: Stillman, F et al. (2003) Evaluation of the American Stop Smoking Intervention Study (ASSIST): A Report of Outcomes. J Natl Cancer Inst. 95 (22): 1681-1691. Mercier, C et al. (2007) Improving Newborn Preventive Services at the Birth Hospitalization: A Collaborative, Hospital-Based Quality-Improvement Project. Pediatrics. 120: 3, pages 481 -488. Work due: Preliminary Research Proposal is due.
Problem set: Applying descriptive and inferential statistical approaches to a sample data set derived from a public health program evaluation project, using SPSS v18. Module 11 Description: Improvement science as applied public health research. What are the types of problems that quality improvement projects address? How is quality improvement used in
5
organizations and communities to change processes and outcomes? What are the data that can be obtained in improvement projects, and how can these be used to support changes that are made in public health systems and settings? Readings: Davidoff F and Batalden P (2005) Toward stronger evidence on quality improvement. Draft publication guidelines: the beginning of a consensus project Qual Saf Health Care; 14:319-325. Instructor PowerPoint presentation.
Module 12 Description: Review of key concepts relating to the goals of public health applied research, possible data sources, original data collection methods, concepts of validity and reliability, descriptive and inferential data analysis, effective communication of data and statistical findings. Discussion of ways that applied research is key to program evaluation and quality improvement projects. Readings: Work: None. Final Research Proposals are due (PowerPoint presentations by students).
(Note: In the semester- long course, students will have an opportunity to read and provide peer review to other classmate’s proposals in the final week of the course. They will utilize these critiques to finalize their proposal.)
6
1
participate. There is no minimum number of times students must participate, but contributions totaling at least 200 words will be expected. Postings will be evaluated for the quality of their content and how well they reflect the readings and/or PowerPoint presentations for each module. Each discussion will be open for one week. Article critiques (20% of final grade): A total of seven articles will be read during the course, and for each article students will write an approximately two page “critique” of the article. Critiques should be written clearly and reflect the students’ learning from the corresponding course module as well as from previous modules. Students will not be graded on spelling or punctuation but rather on how well their writing reflects key ideas from the course and based on their application of critical thinking skills. Problem sets (20% of final grade): A total of five statistical/methodological problem sets will be completed during the course. Each problem set will provide students with background information about an applied public health dataset and the data itself. Each problem set will also describe a specific question that students will answer, using a combination of what they have learned in the corresponding (and previous) modules and the SPSS v18 statistical software package (available for free as a download to UVM students). Students’ work for each problem set will be evaluated based on the quality of a brief (one page or less) written summary of the answers to the questions posed for the problem, as well as copies of the statistical output that supports their answers. Final research proposal (20% of final grade): Starting during Module 4, students will begin developing a sample research proposal that addresses an applied public health research question. For example, a student might write a mock grant proposal to do a study of the effects of fluoridation on a set of dental health measures in a rural community, in which they described the nature of the research question, propose methods for answering the questions and describe how they would collect and analyze data that would allow them to make a strong inference about the applied research question. Proposals will be between seven and ten pages in length, and will be evaluated based on how well the proposal demonstrates mastery of the concepts presented in the course. Readings: The main text for the course is Essentials of Biostatistics in Public Health (2nd edition, 2011) by Lisa Sullivan. Additional readings will consist of journal articles and book chapters, as well as PowerPoint presentations developed by the instructor, and these are listed as part of each module described below. Readings should be completed prior to the start of each module.
Course calendar: Module 1 Description: Overview of course and requirements. What is applied public health research, and how does it differ from other sorts of research? Examples of applied public health research
2
designs. What sorts of questions does applied research in public health answer, and what needs does it meet? Review of studies that represent rigorous applied studies. Readings: Sullivan, Chapter 2. Instructor PowerPoint presentation. Module 2 Description: What are validity and reliability? How do we assess validity and reliability or data collection procedures, instruments, analyses and claims? What are threats to validity and reliability of applied research, and how can these be ameliorated? What are the biases that exist in applied research, and how can these be minimized? Readings: Bowling, A (2002) Research methods in health: investigating health and health services. (Chapter 6, pages 144-153). Levin, S et al. (2003) Physical Activity and Body Mass Index Among US Adolescents. Arch Pediatr Adolesc Med.157:816-820.
Critique of:
Module 3 Description: More detailed exploration of the types of data that are available for applied public health research. Using existing data sets versus original data collection. Issues around data quality and proper data management. Review of how data should be organized for analysis. Readings: Sullivan, Chapter 4. Instructor PowerPoint presentation. Problem set: Identifying threats to data validity and errors in data management in a sample data set. Module 4 Description: Review of qualitative approaches to data collection, including focus (discussion) groups, interviews and others. Identifying appropriate questions and methods for collecting these data. Analyzing qualitative data using less formal versus more formal approaches. Design of effective surveys and interview guides. Description and discussion of the requirement for the Research Proposal course requirement. Readings: Jones J and Hunter D (1995) Qualitative Research: Consensus methods for medical and health services research. BMJ 311 : 376 (Published 5 August 1995) Instructor PowerPoint presentation.
3
Critique of:
Crowe, M et al. (2011) Was it something I did wrong? A qualitative analysis of parental perspectives of their child's bipolar disorder. Journal of Psychiatric and Mental Health Nursing: 18:4, 342–348.
Module 5 Description: Overview of quantitative data sources, including behavioral, biological and environmental data. Outcome versus process measures. Implications of the type and quality of data for the strength of the descriptions/ inferences you are trying to make. Overview of probability distributions. Readings: Sullivan, Chapter 5. Instructor PowerPoint presentation.
Module 6 Description: Descriptive (quantitative) data analysis. What statistics are used for descriptive quantitative research, and what do they tell us? Example descriptive analyses using small data sets and the SPSS software package. Readings: Sullivan, Chapter 6. Instructor PowerPoint presentation. Problem set: Using SPSS v18 to describe a dataset that includes demographic and outcome variables.
Module 7 Description: Inferential (quantitative) data analysis—Part 1. Statistical approaches for small and/or highly variable datasets. Overview of non-parametric versus traditional statistical analysis, and their relative strengths and weaknesses. Analyzing categorical data sets. Reviewing published articles with an eye to the validity of their statistical analysis. Readings: Critique of: Sullivan, Chapter 10. Duncan, P et al. (unpublished). A Statewide Initiative to Support Primary Care Practices in Improving Adolescent Preventive Services.
Problem set: Using SPSS v18 to conduct hypothesis testing using categorical statistical tests such as Fisher’s Exact test and Chi Square.
4
Module 8 Description: Inferential (quantitative) data analysis—Part 2. Overview of statistical approaches for more robust quantitative data sets, including univariate (e.g., t-tests, ANOVA) and multivariate (e.g., logistic regression, MANOVA) analysis. Reviewing published articles with an eye to the validity of their statistical analysis. Readings: Sullivan, Chapters 7 & 9.
Problem set: Using SPSS v18 to conduct univariate hypothesis testing with T-tests, ANOVA and non-parametric (e.g., Kruskall-Wallace) statistical tests.
Module 9 Description: Overview of effective presentation of quantitative data. Why are some data displays more versus less appropriate? Examples of effective data displays, and discussion of how to reduce the potential for confusing consumers of quantitative reports. Readings: Critique of: Instructor PowerPoint presentation. van Heuvelen, M et al. (2005) Differences between participants and nonparticipants in an RCT on physical activity and psychological interventions for older people. Aging Clin Exp Res. 17; 236-245.
Module 10 Description: Program evaluation as applied public health research: goals, methods, analysis and reporting of results. How are program evaluation data different from other data? Review of example program evaluations and discussion of the relative strengths and barriers to effective program evaluation. Readings: Stillman, F et al. (2003) Evaluation of the American Stop Smoking Intervention Study (ASSIST): A Report of Outcomes. J Natl Cancer Inst. 95 (22): 1681-1691. Mercier, C et al. (2007) Improving Newborn Preventive Services at the Birth Hospitalization: A Collaborative, Hospital-Based Quality-Improvement Project. Pediatrics. 120: 3, pages 481 -488. Work due: Preliminary Research Proposal is due.
Problem set: Applying descriptive and inferential statistical approaches to a sample data set derived from a public health program evaluation project, using SPSS v18. Module 11 Description: Improvement science as applied public health research. What are the types of problems that quality improvement projects address? How is quality improvement used in
5
organizations and communities to change processes and outcomes? What are the data that can be obtained in improvement projects, and how can these be used to support changes that are made in public health systems and settings? Readings: Davidoff F and Batalden P (2005) Toward stronger evidence on quality improvement. Draft publication guidelines: the beginning of a consensus project Qual Saf Health Care; 14:319-325. Instructor PowerPoint presentation.
Module 12 Description: Review of key concepts relating to the goals of public health applied research, possible data sources, original data collection methods, concepts of validity and reliability, descriptive and inferential data analysis, effective communication of data and statistical findings. Discussion of ways that applied research is key to program evaluation and quality improvement projects. Readings: Work: None. Final Research Proposals are due (PowerPoint presentations by students).
(Note: In the semester- long course, students will have an opportunity to read and provide peer review to other classmate’s proposals in the final week of the course. They will utilize these critiques to finalize their proposal.)
6
